Short‐Chain Fatty Acids Activate Myenteric Neurons and Delay Colonic Motility via Free Fatty Acid Receptors 2 and 3 in the Mouse

ABSTRACT

Background

Short‐chain fatty acids (SCFAs), metabolic products of the gut microbiota, are important regulators of intestinal homeostasis. SCFAs exert their effects through free fatty acid receptors (FFAR), which are located on enteroendocrine cells and enteric nerves. The actions of SCFAs include the regulation of motility. How SCFAs affect enteric neuronal activity and the regulation of colonic motility remains to be fully elucidated.

Methods

Intact preparations of the proximal colon from mice expressing a genetically encoded fluorescent Ca²⁺ reporter specifically in intrinsic primary afferent neurons (IPANs; Calb1‐GCaMP6 mice) or in all enteric neurons (Wnt1‐GCaMP6 mice) were used. Preparations were luminally perfused with SCFAs, with or without the FFAR2 antagonist, β‐hydroxybutyrate (BHB) or the FFAR3 antagonist, GLPG0976. SCFA‐induced Ca²⁺ responses were visualized by confocal microscopy. Colonic transit was determined using a bead propulsion assay.

Key Results

In Calb1‐GCaMP6 mice, propionate exhibited the highest and most sustained Ca²⁺ fluorescence (∆F/F₀ 288.2 ± 81.2) with the fastest time to peak (7.3 ± 1.6 min) and longest decay time (10.1 ± 3.0 min), followed by acetate, butyrate, and a mixture of the three SCFAs at physiological concentrations. Similar patterns were observed in Wnt‐1‐GCaMP6 mice, where propionate induced the highest Ca²⁺ fluorescence (∆F/F₀ 240.5 ± 25.7) and the longest decay time (10.2 ± 0.9 min). BHB and GLPG0976 significantly attenuated SCFA‐induced Ca²⁺ fluorescence in both IPANs and myenteric neurons. SCFAs delay colonic bead propulsion, an effect abolished by both FFAR antagonists.

Conclusions and Inferences

Acutely, luminal administration of SCFAs activates myenteric neurons and slows colonic transit through FFAR2 and FFAR3 receptors.

Plain Language Summary

The gut microbiota makes factors that influence gut health and the passage of food along the gut. How they do this remains to be well understood. Here we investigated the mechanisms by which some of these factors, called short chain fatty acids, influence the nerves that control movement in the colon. We found that each of the short chain fatty acids we tested exerted distinct stimulatory effects on the activity of nerves mediated by specific receptors. We also found that these short chain fatty acids, at levels normally found in the gut, slows down the movement of the colon. Together our results provide new insights into the mechanisms by which microbial mediators modulate the nervous control and motor function of the colon.

Key Points

• Short‐chain fatty acids (SCFAs) are important for gut homeostasis. How SCFAs affect enteric neuronal activity and regulate colonic motility remains to be fully elucidated.

• Luminal administration of physiological concentrations of acetate, propionate, butyrate, and their combination activated myenteric neurons with distinct response patterns via both free fatty acid receptor (FFAR)2 and FFAR3. The combination of SCFAs slows colonic transit via FFAR2 and FFAR3.

• Together, these data provide new insights into the mechanisms by which these microbial mediators modulate enteric neuronal function and colonic motility.

Short‐chain fatty acids (SCFAs), including propionate, acetate, and butyrate, activate IPANs and myenteric neurons and delay colonic motility via FFAR2 and FFAR3. GLPG0976 and β‐hydroxybutyrate (BHB) were used as selective antagonists for FFAR2 and FFAR3, respectively, to confirm receptor‐specific effects.

Abbreviations

5‐HT

serotonin

ENS

enteric nervous system

FFAR

free fatty acid receptor

IPANs

intrinsic primary afferent neurons

SCFA

short chain fatty acid

1. Introduction

Short‐chain fatty acids (SCFAs) are the primary metabolic products of microbial fermentation of nondigestible poly‐ and oligosaccharides. The three major SCFAs, acetate, propionate, and butyrate, are present in a ratio of approximately 3:1:1, respectively, in the lumen of the gut [1, 2], and in the proximal colon are present at a collective concentration of approximately 100 mM [1, 2]. Lower concentrations are present in the distal colon, liver, and bloodstream [3, 4]. Short‐chain fatty acids mediate a variety of functions in the gut, notably, providing an energy source for enterocytes, modulation of gut motility, regulation of immune responses, gut hormone secretion, anion secretion, inflammation and the maintenance of intestinal barrier function [5, 6, 7, 8, 9, 10, 11, 12, 13].

The diverse actions of SCFAs are primarily mediated through their interaction with specific G‐protein coupled receptors: free fatty acid receptor 2 (FFAR2/GPR43) and free fatty acid receptor 3 (FFAR3/GPR41) [14]. FFAR2 is G_q‐ and G_i/o‐coupled and is predominantly expressed on intestinal epithelial cells, colonic lamina propria cells, adipocytes, and skeletal muscle [14, 15]. Conversely, FFAR3 is G_i/o‐protein‐coupled and expressed on enteric neurons, adipocytes, lymph nodes, bone marrow, and peripheral mononuclear cells [11, 14, 15, 16]. Through these receptor‐mediated pathways, SCFAs can influence specific physiological processes at the level of epithelium, including serotonin (5‐HT) release from EC cells [17], modulation of hormone‐secreting L cells [18], and regulation of epithelial secretory functions [7, 9, 10, 19, 20]. Interestingly, previous studies suggest the effect of SCFAs on gut motility is both FFAR2 and FFAR3‐dependent [8, 10, 21] or independent [22].

Colonic motility is regulated by the enteric nervous system (ENS), a complex network of neurons and glia embedded in the wall of the gut [23]. It is composed of two plexuses, the submucosal plexus and myenteric plexus, which control secretion and motility [23]. Within the ENS, intrinsic primary afferent neurons (IPANs) are the first neurons to respond to luminal stimuli [23, 24, 25]. Integration of this information at the level of the myenteric plexus ensures the appropriate coordination of contractility and secretion to facilitate the movement of luminal contents [26]. SCFAs have been shown to both activate and inhibit myenteric neurons, regulate the phenotype of myenteric neurons, and control motor and secretory responses [1, 7, 13, 21, 27, 28, 29, 30, 31, 32]. How luminal SCFAs affect enteric neuronal activity and regulate colonic motility still remains to be fully elucidated.

Here we investigated the effects of SCFAs in myenteric neuronal activity and the mechanisms by which SCFAs influence enteric neuronal activation and their contribution to colonic motility. We used intact preparations of the proximal colon of Calb1‐GCaMP6 and Wnt1‐GCaMP6 mice, which express a Ca²⁺ reporter specifically in IPANs and in all enteric neurons, respectively [33, 34]. We found that luminal administration of physiological concentrations of acetate, propionate, butyrate, and their combination increased Ca²⁺ fluorescence in IPANs and myenteric neurons with distinct response patterns. The combination of SCFAs increased Ca²⁺ fluorescence in enteric neurons via both FFAR2 and FFAR3. We also found that the combination of SCFAs slows colonic transit via FFAR2 and FFAR3. These results provide new insights into the mechanisms by which SCFAs modulate enteric neuronal function and colonic motility.

2. Materials and Methods

2.1. Ethics Statement

This study was approved by the Health Sciences Animal Care Committee of the University of Calgary (animal use protocols AC23‐0119, AC23‐0135) following the guidelines of the Canadian Council on Animal Care and adheres to the Animals in Research: Reporting In Vivo Experiments (ARRIVE) guidelines.

2.2. Animals

Adult male and female mice (aged 10–20 weeks) were purchased from The Jackson Laboratory (Bar Harbor, ME). B6J.Cg‐Gt(ROSA)26Sor^{tm96(CAG‐GCaMP6s)Hze}/MwarJ mice (Jackson 028866) were bred at the University of Calgary and crossed with either Tg(Wnt1‐cre)11Rth Tg(Wnt1‐GAL4)11Rth/J (Jackson 003829) or B6.Cg‐Calb1^{tm2.1(cre)Hze}/J mice (Jackson 028532) to obtain Wnt1‐GCaMP6 or Calb1‐GCaMP6 mice, respectively.

Adult male and female mice (aged 8–20 weeks) were group‐housed (1–5 mice per cage) under a 12 h light–12 h dark cycle (lights off at 19.00 h) in a temperature‐ and humidity‐controlled room. Mice were allowed free access to a standard diet (LabDiet, St Louis, MO, USA; 5061 or 5062) and tap water. Animals were euthanized by cervical dislocation under deep anesthesia (4% isoflurane in oxygen). A 4–5 cm segment of proximal colon was collected and placed in Krebs buffer immediately after euthanasia. Colonic segments were cleaned of luminal content by Krebs buffer before performing live cell imaging experiments.

2.3. Live Cell Confocal Imaging

Live cell imaging and calcium fluorescence analysis were conducted as previously described [33, 34]. Briefly, the intact proximal colon from Wnt‐1‐GCaMP6 mice, which specifically express a genetically encoded fluorescent calcium reporter in enteric neurons and glia, or Calb1‐GCaMP6 mice, which specifically express a genetically encoded fluorescent calcium reporter in IPANs, were placed a custom‐made perfusion chamber. The intact proximal colon was connected via VMR silicone tubing with diameter 0.25 in. (Avantor, Canada; cat#89068–484) to an infusion and withdrawal pump at a flow rate of 10 mL.h⁻¹. Tissues were bathed in Krebs solution containing (in mM): 121 NaCl, 5.9 KCl, 2.5 CaCl₂, 1.2 MgCl₂, 1.2 NaH₂PO₄, 14.4 NaHCO₃, 11 glucose (Sigma, St. Louis, MO) with 3 μM nicardipine hydrochloride (Sigma, cat #N7510) and 1 μM scopolamine (Sigma; cat #S1875) to inhibit smooth muscle contractions, as previously described [33, 35].

For perfusion experiments, the proximal colon was perfused with Krebs solution under physiological conditions (pH~7.4, 5% CO₂, 95% O₂ and 37°C) during the equilibration period. The intraluminal flow rate and temperature were maintained constant for the duration of the experiment using PHD 2000 syringe pumps (Harvard Apparatus, Holliston, MA) combined with 1LG Syringe heaters (New Era Pump Systems Inc., Farmingdale, NY). The flow and temperature in the bath were maintained by a Minipuls 3 peristaltic pump (Gilson, Middleton, WI) and a TC‐344B heater controller (Harvard Apparatus, Holliston, MA). SCFAs were perfused into the luminal or serosal side of the proximal colon for 3 min. The concentration of SCFAs was chosen based on previous studies [1]. At the end of each experiment, neuronal responsiveness was confirmed by adding 75 mM K⁺ to the bathing solution.

The visualization of intracellular Ca²⁺ fluorescence intensity in myenteric neurons and IPANs was obtained by live‐cell confocal microscopy (A1R HD25 confocal system, combined with a Ti2‐Eclipse inverted microscope, Nikon, Tokyo, Japan). The 3D video footage was stabilized, quantified, and analyzed for individual neurons using Imaris software versions x64 9.7.2 to 10.1.1 (Bitplane, Belfast, UK).

In Wnt1‐GCaMP6 mice, 15–35 neurons (determined by their size) in 2–3 ganglia were randomly selected per experiment. Data containing x–y coordinates and mean fluorescence intensities for each neuron were exported from Imaris to a spreadsheet. We then calculated and plotted the fluorescence changes for each cell over time. To account for minor x‐y displacement caused by the intrinsic contractility of the intact colon, we utilized the Imaris spot‐tracking algorithm, ensuring a consistent region of interest for each neuron throughout the recording. The initial fluorescence (F₀) is indicative of the mean fluorescence intensity just before the reaction to a stimulus, while the maximal fluorescence intensity (F_max) represents the highest fluorescence intensity reached by a neuron after a stimulus. The difference in fluorescence (ΔF) is equal to F_max−F₀. Individual neurons were considered responsive to a stimulus when their ΔF/F₀ was greater than 20%. A ΔF/F₀ < 20% was considered noise. Neurons that did not respond to KCl were excluded from this study. Time to peak is the amount of time from (F₀) to F_max. Decay time is the time from F_max to ΔF/F₀ reaching zero. Proximal colon was used in this study because it serves as a primary site for studying SCFA–host interactions [2, 4, 21, 36, 37].

For Calb1‐GCaMP6 mice, approximately 3–5 neurons in up to five ganglia per experiment were used. The result from Calb1‐GCaMP6 mice potentially contains both plexuses because it is not possible to distinguish between submucosal and myenteric plexuses in these preparations.

2.4. SCFA and Drug Treatments

Preparations were treated with acetate (Sigma; cat #S2889; 60 mM), propionate (Sigma; cat #P1880; 20 mM) or butyrate (Sigma; cat #303410; 20 mM) or a SCFA mixture of acetate, propionate, and butyrate (100 mM total). In order to determine whether FFARs were involved in the response to SCFAs, some preparations were co‐treated with the FFAR2 antagonist GLPG0976 (Cayman, MI, cat #28108; 0.5 μM [38, 39, 40]), or FFAR3 antagonist β‐hydroxybutyrate (BHB; Sigma, MO, USA, cat #H6501; 100 μM [41, 42]), and the SCFA mixture. In additional experiments, the SCFA mixture (375 μM) was added to the serosal, the concentration present in the circulation [4, 43]. In all SCFA experiments, osmolarity was adjusted by reducing NaCl. The pH of all solutions was adjusted to ~ pH 7.4.

2.5. Colonic Bead Propulsion

Mice (not fasted) were briefly anesthetized with 2% isoflurane in oxygen during the enema and bead insertion procedures. The enema (0.1 mL) was administered using a volume previously confirmed to reach the proximal colon [34]. Subsequently, mice received an enema (0.1 mL) containing a 100 mM SCFA mixture, either alone or in combination with GLPG0976 (0.5 μM) or BHB (100 μM). Some groups received GLPB0976 or BHB alone. 20, 60, and 120 min after the initial enema, mice were very briefly anesthetized again, and a small glass bead (2 mm diameter) was inserted 2 cm into the anus. Mice recovered in less than 1 min after bead insertion and were placed in a large glass beaker and left until the bead was expelled. The time for the bead to be expelled was recorded in each case, and after the last bead was expelled, mice were euthanized.

2.6. Statistics

Data were analyzed using paired or unpaired t tests, one‐way ANOVA with Dunnett's or Tukey's post hoc tests or two‐way ANOVA, followed by a Tukey's test, as appropriate, using GraphPad Prism version 10 (Boston, MA, USA). A p‐value of < 0.05 was considered statistically significant in all cases.

3. Results

3.1. Luminal SCFAs Activate IPANs

Previous studies have shown that IPANs in isolated tissues preparations respond to one or more SCFAs [30, 44]. We first confirmed that in intact, full‐wall thickness preparations IPANs responded to SCFAs. The degree of Ca²⁺stimulation, from highest to lowest, was propionate (20 mM, Ca²⁺ intensity = 377.0 au; ∆F/F₀ = 288.2 ± 81.2; 18 neurons, n = 5), acetate (60 mM, Ca²⁺ intensity = 318.5 au; ∆F/F₀ 197.0 ± 26.8; 11 neurons, n = 4), butyrate (20 mM, Ca²⁺ intensity = 168.1 au; ∆F/F₀ 129.6 ± 23.1; 15 neurons, n = 5), and the mixture of SCFAs (Ca²⁺ intensity = 345.3 au; ∆F/F₀ 216.6 ± 56.13; 8 neurons, n = 4), respectively (Figure 1A–C,F; Figure 2; Videos [Link], [Link]).

FIGURE 1.

Confocal imaging of Ca²⁺ fluorescence in the intact proximal colon of Calb1‐GCaMP6 mice, stimulated by acetate, propionate, butyrate or the SCFA mixture. (A) Intensity of Ca²⁺ fluorescence in individual neurons under baseline conditions and after luminal perfusion with acetate (60 mM; n = 4), propionate (20 mM; n = 5), butyrate (20 mM; n = 5) or SCFA mixture (100 mM; n = 4). Each symbol represents 1 neuron. Data were compared using a paired t test. (B) Magnitude of ΔF/F₀ in individual neurons in response to luminal perfusion with acetate, propionate, butyrate or SCFAs mixture. Each symbol represents 1 neuron. (C) Magnitude of ΔF/F₀ in individual animals (averaging the data from individual neurons) in response to luminal acetate, propionate and butyrate or SCFA mixture. (D) Magnitude of ΔF/F₀ in individual neurons in response to luminal SCFA mixture in the presence or absence of β‐hydroxybutyrate (BHB; 100 μM) or GLP0974 (0.5 μM). Data were compared using Dunnett's ANOVA. (E) Magnitude of ΔF/F₀ in individual animals (averaging the data from individual neurons) in response to luminal SCFAs mixture with BHB or GLP0974. Data were compared using Dunnett's ANOVA. (F) Representative fields of view of the intrinsic primary afferent neurons in the proximal colon at baseline and after luminal perfusion with acetate, propionate, butyrate or SCFA mixture in the presence and absence of BHB or GLP0974. Scale bar: 20 μm.

FIGURE 2.

Representative traces of Ca²⁺ fluorescence in individual neurons in response to luminal perfusion of SCFAs in Calb1‐GCaMP6 mice. Ca²⁺ fluorescence in individual neurons in response to luminal acetate (60 mM), propionate (20 mM), butyrate (20 mM) or the SCFA mixture (100 mM) in the presence and absence of BHB or GLP0974. At the end of each experiment, preparations were treated with KCl to test the cell viability. Darkened color above the trace indicates the time of luminal administration of the SCFAs or the addition of KCl to the bath.

The percentage of neurons responding to acetate, propionate, butyrate, and the SCFA mixture was similar between the groups (100.0% ± 0% acetate, 94.4% ± 24.8% propionate, 100% ± 0% butyrate, and 87.5% ± 10.0% SCFA mixture, respectively) (Table S1A). Together, these data indicate that propionate has the most potent and sustained effect on the activity of IPANs based on the elevation of intracellular Ca²⁺.

We next examined whether the FFAR2 and FFAR3 were involved in the response to the SCFA mixture. The Ca²⁺ elevation stimulated by the SCFA mixture was significantly reduced by BHB and GLPG0976, with 7.7 ± 25.0 and 15.4% ± 28.9% of neurons responding, respectively, suggesting that luminal SCFAs increase Ca²⁺ activity in IPANs via both FFAR2 and FFAR3 (Figure 1D,E; Table S1A; Videos S5 and S6).

Luminal propionate had the fastest time to peak (TTP, 7.3 ± 1.6 min) followed by the SCFAs mixture (9.6 ± 3.2 min), butyrate (9.8 ± 2.3 min), and acetate (10.7 ± 2.3 min) in 7, 7, 13, and 10 neurons, respectively. The slowest decay time (DT) was propionate (10.1 ± 3.0 min), followed by SCFA mixture (9.9 ± 4.6 min), butyrate (8.9 ± 1.7 min), and acetate (4.7 ± 3.3 min), respectively (Figure 3A,B).

FIGURE 3.

Quantification of time to peak and decay time in Calb1 GCaMP6 mice. (A) Time to peak in individual neurons in response to luminal perfusion of acetate (60 mM), propionate (20 mM), butyrate (20 mM) or the SCFA mixture (100 mM). (B) Decay time in individual neurons responding neurons to luminal perfusion of acetate, propionate, butyrate or the SCFA mixture. Each symbol represents 1 neuron.

3.2. Luminal SCFAs Stimulate Myenteric Neurons

We next investigated the effect of luminal administration of individual SCFAs and the SCFA mixture on neuronal activity in the colonic myenteric plexus as a whole, since we would expect that reflex activity would be initiated by the IPANs. In Wnt1‐GCaMP6 mice, luminal treatment with acetate (60 mM), propionate (20 mM), butyrate (20 mM), or a mixture of the three SCFAs (100 mM) significantly increased the intensity of Ca²⁺ fluorescence compared to baseline (Figure 4A–C; Figure 5; Videos [Link], [Link]). The greatest increase in mean Ca²⁺ fluorescence was observed after propionate treatment (Ca²⁺ intensity = 279.1 au; ∆F/F₀ = 240.5 ± 25.7; 90 neurons, n = 5), followed by acetate (Ca²⁺ intensity = 245.0 au; ∆F/F₀ = 213.7 ± 11.8; 79 neurons, n = 5) and butyrate (Ca²⁺ intensity = 261.6 au; ∆F/F₀ = 199.3 ± 11.9; 78 neurons, n = 7), respectively. Surprisingly, although the SCFA mixture increased Ca²⁺ fluorescence, it neither had an additive or synergistic effect compared to the individual SCFAs (Ca²⁺ intensity = 331.0 au; ∆F/F₀ = 245.7 ± 18.8; 102 neurons, n = 9) (Figure 4A–C; Figure 5). This pattern of activity was similar to that observed for activation of IPANs.

FIGURE 4.

Confocal imaging of Ca²⁺ fluorescence in the intact proximal colon of Wnt1‐GCaMP6 mice, stimulated by acetate, propionate, butyrate and the SCFA mixture. (A) Intensity of Ca²⁺ fluorescence in individual neurons under baseline and luminal perfusion with acetate (60 mM; n = 5), propionate (20 mM; n = 5), butyrate (20 mM; n = 7) or SCFAs mixture (100 mM; n = 7). Each symbol represents 1 neuron. Data were compared using a paired t test. (B) Magnitude of ΔF/F₀ in individual neurons in response to luminal perfusion with acetate, propionate and butyrate or the SCFA mixture. Each symbol represents 1 neuron. (C) Magnitude of ΔF/F₀ in individual animals (averaging the data from individual neurons) in response to luminal acetate, propionate and butyrate or the SCFA mixture. (D) Magnitude of ΔF/F₀ in individual neurons in response to luminal SCFAs mixture with β‐hydroxybutyrate (BHB; 100 μM) or GLP0974 (0.5 μM). Data were compared using Dunnett's ANOVA. (E) Magnitude of ΔF/F₀ in individual animals (averaging the data from individual neurons) in response to luminal SCFAs mixture with BHB or GLP0974. Data were compared using Dunnett's ANOVA. (F) Representative fields of view of the myenteric neuron in the proximal colon at baseline and after luminal acetate, propionate, butyrate or the SCFA mixture in the presence and absence of BHB or GLP0974. Scale bar: 100 μm.

FIGURE 5.

Representative traces of Ca²⁺ fluorescence in individual neurons in response to luminal SCFAs in Wnt1‐GCaMP6 mice. Ca²⁺ fluorescence in individual neurons in response to luminal acetate (60 mM), propionate (20 mM), butyrate (20 mM), or the SCFA mixture (100 mM) in the presence and absence of BHB (100 μM) or GLP0974 (0.5 μM). At the end of each experiment, preparations were treated with KCl to test cell viability. Darkened color above the trace indicates the time of luminal acetate, propionate, butyrate, or the SCFA mixture or the addition of KCl to the bath.

All neurons studied responded to luminal acetate, propionate, butyrate, and the combined SCFAs (Table S1B). Conversely, serosal treatment of the SCFA mixture at concentrations found in the circulation (375 μM) induced only a very low level of Ca²⁺ fluorescence (∆F/F₀ = 42.2 ± 6.6; n = 3) in ~70% of the 91 neurons examined. Due to the low Ca²⁺ response observed with serosal SCFA treatment, we did not investigate its effects further.

Since FFAR2 and FFAR3 have been shown to regulate gut motility and are expressed on enteric neurons and enteroendocrine cells [16, 21], we hypothesized that SCFAs stimulated enteric neuronal activity via one or both of these receptors. Co‐perfusion of BHB, a FFAR2 antagonist or GLPG0974, a FFAR3 antagonist, with the mixture of SCFAs showed that both FFAR2 and FFAR3 antagonists significantly reduced the effect of SCFA‐induced Ca²⁺ fluorescence in myenteric neurons with 47.4 ± 21.1 and 54.3% ± 8.0% of neurons responding, respectively (Figure 4D–F; Figure 5; Videos S11 and S12), suggesting that SCFAs stimulated myenteric neurons via both FFAR2 and FFAR3.

Next, we calculated the TTP and DT of SCFAs‐induced Ca²⁺ fluorescence in myenteric neurons. Luminal propionate had the fastest TTP (3.6 ± 0.4 min, 72 neurons), followed by acetate (8.8 ± 0.8 min, 66 neurons), SCFA mixture (7.4 ± 0.6 min, 78 neurons), and butyrate (10.0 ± 0.9 min, 76 neurons), respectively. The slowest DT was in response to propionate (10.2 ± 0.9 min, 68 neurons) followed by butyrate (8.5 ± 1.0 min, 39 neurons), SCFA mixture (7.6 ± 0.5 min, 76 neurons), and acetate (5.6 ± 0.8 min, 51 neurons), respectively (Figure 6). This indicates that propionate has the most potent and sustained effect on the activation of myenteric neurons based on the intensity of intracellular Ca²⁺ fluorescence.

FIGURE 6.

Quantification of time to peak and decay time in Wnt1‐GCaMP6 mice. (A) Time to peak in individual neurons in responding neurons after luminal acetate (60 mM), propionate (20 mM), butyrate (20 mM) or the SCFA mixture (100 mM) perfusion (B) Decay time in individual neurons responding neurons after luminal acetate, propionate, butyrate or SCFAs mixture perfusion. Each symbol represents 1 neuron.

We considered the possibility that sex differences of the mice could influence activation of enteric neurons. The levels of Ca²⁺ fluorescence of individual neurons of Wnt1‐GCaMP6 mice showed that male mice treated with luminal acetate, propionate, and butyrate had significantly higher Ca²⁺ levels compared to female mice. Curiously, there was no statistically significant sex difference in the response to the SCFA mixture (Figure 7A–D).

FIGURE 7.

Sex difference in Ca²⁺ fluorescence in Wnt‐1 GCaMP6 mice. (A–D) Magnitude of ΔF/F₀ in individual neurons in response to luminal acetate (60 mM), propionate (20 mM), butyrate (20 mM), or the SCFA mix (100 mM), in male and female mice. Data were compared using unpaired t tests.

To test whether any mechanical stimulus caused by intraluminal perfusion with Krebs buffer affected the Ca²⁺ fluorescence, we compared the level of Ca²⁺ intensity at the time point corresponding to the peak responses (~5–7 min) to baseline in myenteric neurons (Figure S1). No significant change in the level of Ca²⁺ fluorescence was detected between baseline and perfusion, confirming that mechanical stimulation did not contribute to luminal SCFA stimulation of neuronal Ca²⁺ fluorescence.

3.3. Luminal SCFAs Delay Colonic Transit

Finally, we explored the effect of the SCFA mixture on colonic motility. Using the bead propulsion assay, we found that an enema of the SCFA mixture significantly increased expulsion time at 20 min compared to Krebs enema control (p < 0.0001) (Figure 8A). This effect was abolished by both BHB (p = 0.0014) and GLPG0976 (p = 0.0176), suggesting that luminal SCFAs delayed colonic motility via both FFAR2 and FFAR3. We observed that the male mice had slower colonic transit time compared to female mice in the response (p = 0.0037) to the SCFA mixture (Figure 8B). The effect of SCFAs on colonic motility was relatively transient, and no effects were observed on bead propulsion at 60 or 120 min after treatment (data not shown).

FIGURE 8.

The effects of SCFAs on colonic motility. (A) Male and female mice received an enema of physiological saline (male n = 5, female n = 4), SCFAs (male n = 8, female n = 7), SCFAs+100 μM BHB (male n = 5, female n = 5), SCFAs+0.5 μM GLPG0974 (male n = 6, female n = 4), BHB alone (male n = 5, female n = 2), or GLPG0974 alone (male n = 6, female n = 3). Glass bead insertion was performed at 20 after treatment. SCFAs significantly delayed colonic motility after treatment. BHB and GLPG0974 abolished this effect. All groups were compared using two‐way ANOVA with Tukey's multiple comparison test. (B) Male or female mice received an enema of physiological saline (male n = 5, female n = 4), SCFAs (male n = 8, female n = 7). Glass bead insertion was performed at 20 min after treatment. All groups were compared using one‐way ANOVA with Tukey's multiple comparison test.

4. Discussion

Here we confirm and extend studies that have demonstrated SCFAs activate myenteric neurons in full‐wall thickness preparations of the proximal colon [1], providing additional insights into how luminal SCFAs affect enteric neuronal activity and colonic motility. We demonstrate using an enema that physiological concentrations of luminal SCFAs (acetate, butyrate, or propionate) stimulate IPANs and myenteric neurons via FFAR2 and FFAR3, with different patterns of neuronal responses. In vivo, an enema of SCFAs transiently slows colonic motility; effects that are also mediated by FFAR2 and FFAR3.

Physiologically, three major types of SCFAs, acetate, propionate, and butyrate, are present in the colonic lumen at an approximately 3:1:1 ratio. Maintaining this ratio is important for preserving gut homeostasis and regulating motility [1, 2]. The composition and metabolic activity of the gut microbiota influence this ratio by determining how specific precursor substrates are broken down. For example, Bacteroides thetaiotaomicron produces propionate that supports the balance of gut intestinal goblet cell differentiation and ameliorates colitis in mice treated with dextran sodium sulfate [45, 46]. In contrast, excessive propionate production that is often driven by the accumulation of its precursors succinate and lactate, and facilitated by bacteria such as Veillonella parvula , can suppress butyrate production. This shift in SCFA ratio increases pro‐inflammatory markers and reduces anti‐inflammatory markers in inflammatory bowel disease and irritable bowel syndrome [12, 46, 47]. These findings highlight the importance of using appropriate physiological ratios and concentrations of SCFAs when studying their effects on colonic motility.

To investigate the effect of SCFAs on enteric neurons, Fung et al. [1] used full‐wall thickness proximal colon preparations from transgenic mice expressing GCaMP3 as a model. This study showed that luminal SCFAs do not directly activate myenteric neuronal responses but rather require the mucosa to initiate the potential effect of SCFAs on neuronal activity [1]. These findings are consistent with previous work showing that the effects of the SCFA propionate on colonic motility in the rat required an intact mucosa and were sensitive to tetrodotoxin and 5‐HT₄ receptor antagonists, showing they were both neurally mediated, but this occurred after release of 5‐HT from enterochromaffin cells [13, 31]. Together, this suggests that SCFA‐mediated motility effects may primarily be mediated by mucosal enteroendocrine signaling rather than direct neuronal activation. Supporting this idea, in response to chemical, mechanical, and optical stimulation applied luminally, enterochromaffin cells can release 5‐HT, which plays a key role in initiating colonic motility [17, 48, 49, 50]. These mucosal signals are thought to activate downstream components of the ENS, particularly IPANs, which are central regulators of colonic motility through their control of smooth muscle contraction and relaxation [23]. IPANs serve as critical sensors of luminal contents including microbial products and nutrients, initiating the neural reflexes that regulate motility [23, 24, 26].

Our results showed that luminal propionate activated myenteric neurons to the greatest extent. This is supported by the peak fluorescence, the fastest time‐to‐peak, and the slowest decay time compared to acetate and butyrate. Interestingly, the mixture of the three SCFAs did not demonstrate additive or synergistic effects on Ca²⁺ activity. Propionate is a stable end product of bacteria fermentation and is not cross‐metabolized by the microbiota, unlike acetate, which is frequently utilized as a substrate for butyrate synthesis. It is chemically more stable and remains in a bioavailable form longer than acetate and butyrate [12, 51]. This suggests that propionate may exert stronger effects due to its direct action on host tissues and independent of microbial processing. Since propionate is a potent activator of IPANs, the sensory “first responders” to luminal stimuli of the ENS, its ability to rapidly and strongly trigger these neurons might lead to a more robust downstream activation of the entire myenteric plexus compared to other SCFAs. However, since we cannot rule out the presence of some intestinal bacteria or microbial activity in our setting because we used conventionally housed mice, future studies are required to investigate this possibility.

Previous studies showed that SCFAs can exert their effects via FFARs in a chain‐length‐dependent manner, with propionate (C3) being the most potent agonist for both FFAR2 and FFAR3, acetate (C2) showing greater selectivity for FFAR2, and butyrate (C4) exhibiting stronger activity on FFAR3 [52, 53]. As we used the maximal physiological concentrations of SCFAs, the effect of propionate may predominate over other SCFAs in mixtures, potentially explaining the absence of either additive or synergistic responses in neuronal activation when SCFAs are combined. Because we used a full‐wall thickness preparation of the colon, we could not distinguish the site of action of the SCFAs or the location of the receptors involved. However, based on the fact that FFAR3 is found on enteric neurons [16, 21], and FFAR2 on enteroendocrine cells [13, 16, 54, 55], it seems likely that these are cellular sites of action of the SCFAs in this study. Future studies with cell‐specific receptor knockouts are warranted to address this directly.

SCFAs have been shown to influence colonic motility. An in vitro and ex vivo study demonstrated that apical propionate slowed colonic transit [10, 56]. In contrast, butyrate has been shown to increase colonic transit and the proportion of choline acetyltransferase‐positive enteric neurons following hours to days of exposure [28]. These effects likely reflect adaptive changes in enteric neuronal phenotype and neuromuscular transmission rather than the acute neuronal activation observed in the present study. Hence, the rapid and slower effects of SCFAs on colonic motility may be mediated by distinct signaling mechanisms and neuroplastic alterations within the ENS, reinforcing the differential roles of SCFAs in enteric neuronal function. A previous study reported that acetate reduced contractile frequency in the proximal colon of rats [37] and guinea pigs [2], but another found no effect [31], whereas butyrate enhanced contractile activity and propionate abolished it completely [2]. In rats, a combination of SCFAs (acetate, propionate and butyrate) reduced colonic myoelectrical activity and prolonged transit time [57], supporting the inhibitory effects of SCFAs on colonic motor function observed in this study.

Although our study focused on the proximal colon, SCFAs are present throughout the colonic lumen and may similarly activate enteric neurons in the distal colon. Activation of distal colonic enteric circuits by SCFAs may contribute to coordinated propulsion and overall colonic transit. Therefore, further studies will be necessary to investigate regional differences in SCFA sensitivity along the colon and to determine the relative contributions of proximal versus distal enteric circuits to whole‐organ motility.

A limitation of this study is that we did not identify the specific populations of myenteric neurons that responded to SCFAs beyond IPANs (which are cholinergic) [23]. However, in previous work, the proportion of neuronal nitric oxide synthase and calretinin‐positive neurons responded equally following propionate, acetate, and butyrate treatments [1], suggesting that many subclasses of neurons are activated by SCFAs and the net effects observed functionally rely on the integration of these luminal signals with ongoing neural activity in the plexus. Interestingly, our findings are similar to Kaji et al. [21], who demonstrated that in the rat proximal colon, FFAR3 is expressed on enteric cholinergic neurons. They showed that FFAR3 activation triggers a G_i/o signaling pathway that suppresses cholinergic neural reflexes, providing a plausible mechanistic basis for the inhibitory effects of propionate on colonic motility.

The effect of SCFAs on motility is likely mediated through signaling pathways involving FFAR activation. FFAR2 and FFAR3 play crucial roles in delayed motility [10]. FFAR2 signals through both G_i/o and the pertussis toxin‐insensitive G_q protein families, while FFAR3 acts exclusively through the pertussis toxin‐sensitive G_i/o family [58, 59, 60]. Their stimulation increases intracellular Ca²⁺ levels in many cell types, including enteroendocrine cells [11, 16], sympathetic neurons [41], and enteric neurons and smooth muscle cells involved in regulating colonic motility [8, 10, 21]. Although enteric glial cells are also known to play modulatory roles in enteric signaling and can respond to neurotransmitter release [61], we did not observe an increase in glial Ca²⁺ activity following luminal SCFA perfusion (unpublished observations).

We found that male mice are more sensitive to acetate, propionate, and butyrate with respect to neuronal activity, and to the SCFA mixture with respect to colonic motility. It is possible that male mice have higher sensitivity for cAMP and have hormonal or metabolic factors that modulate SCFA‐induced neuronal and motility responses of our findings. However, a more rigorous study of the sex differences in FFAR expression and other factors is required to explore the mechanisms further.

In conclusion, our study contributes to and extends the evidence that SCFAs regulate enteric neuronal activity and colonic motility. The observed sex differences in SCFA responsiveness suggest the need for sex‐specific approaches in targeting SCFA signaling for the treatment of gastrointestinal disorders such as diarrhea, inflammatory bowel disease, and irritable bowel syndrome. Future studies should further elucidate the mechanistic pathways linking FFAR activation to enteric neuronal circuits and explore how SCFAs influence gut motility in disease models.

Author Contributions

P.W., W.K.M. and K.A.S. designed the studies; P.W., D.L. and L.E.W. conducted experiments and performed data analyses; P.W. and K.A.S. drafted the manuscript. All authors had access to the study data and critically reviewed manuscript drafts and approved the final manuscript for submission. W.K.M. and K.A.S. obtained funding for the study and provided study supervision.

Funding

This work was supported by grants from the Canadian Institutes of Health Research (WKM [PJT‐153290] and KAS [FDN148380]), the Natural Sciences and Engineering Research Council of Canada (WKM [RGPIN/04321–2018]). PW was the recipient of a TRIANGLE Canada fellowship award.

Conflicts of Interest

The authors declare no conflicts of interest.

Supporting information

Video S1: nmo70299‐sup‐0001‐video1.mov.

Video S2: nmo70299‐sup‐0002‐video2.mov.

Video S3: nmo70299‐sup‐0003‐video3.mov.

Video S4: nmo70299‐sup‐0004‐video4.mov.

Video S5: nmo70299‐sup‐0005‐video5.mov.

Video S6: nmo70299‐sup‐0006‐video6.mov.

Video S7: nmo70299‐sup‐0007‐video7.mov.

Video S8: nmo70299‐sup‐0008‐video8.mov.

Video S9: nmo70299‐sup‐0009‐video9.mov.

Video S10: nmo70299‐sup‐0010‐video10.mov.

Video S11: nmo70299‐sup‐0011‐video11.mov.

Video S12: nmo70299‐sup‐0012‐video12.mov.

Data S1: nmo70299‐sup‐0013‐FigureS1‐TableS1.docx.

Figure S1: No effect of mechanical stimulation on myenteric neuronal activation. Ca²⁺ fluorescence at baseline compared to Krebs perfusion in Wnt1‐GCaMP6 mice. Intensity of Ca²⁺ in individual neurons in response to luminal perfusion with Krebs. Each symbol represents 1 neuron (n = 3).

Table S1: Neuronal responses to luminal SCFAs in the proximal colon of Wnt1‐GCaMP and Calb1‐GCaMP mice.

Data Availability Statement

The data that support the findings of this study are available from the corresponding author upon reasonable request.